Is it possible to have voice communication in a room to a room?
I am from max.
I heard from max that one avatar can connect to two rooms, primary and secondary.
What I am asking is to receive the voices of the avatars in the entire room(A) and send them to the avatar(object) in room(B) and vice versa, mixing the voices of the entire room(B),
I would like to mix the voices of the avatars or objects in room(A) and vice versa and have them speak from the avatars or objects in room(B).
11 Replies
Hi @Shigy. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:
----------------------
Based on the information provided in the knowledge sources, Normcore does support high-quality, low-latency voice chat through the
RealtimeAvatarVoice component. This component works on any realtime prefab as long as the RealtimeView component has an owner. It uses the ownership of the RealtimeView to determine if it should be sending audio for this prefab or receiving audio to play back. If the component is owned locally, a microphone stream is created, and if it is owned by a remote client, an AudioOutput component is created, in order to play back audio (source (https://normcore.io/documentation/realtime/xr/voicechat)).
However, the documentation does not provide specific information on whether it's possible to mix voices from different rooms and have them speak from avatars or objects in another room. It does mention that you can connect to multiple rooms simultaneously, which can be used to split large MMORPG spaces into smaller shards or to group traffic to host large events like concerts (source (https://normcore.io/documentation/essentials/common-questions#can-i-connect-to-multiple-rooms-at-the-same-time)). But it doesn't specify if the audio streams can be shared or mixed between these rooms.
For more specific use cases like this, it might be best to reach out to the Normcore team directly or check their more detailed documentation.
----------------------
Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:
----------------------<@1270906710730473525> kapa.ai is still learning and improving, please let us know how it did by reacting below
You could create a voice-only room (Room C in this case) and have only the players that you want to be able to talk to each other connect that way
For one-on-one conversations, it would be possible to do it the way Max indicated
However, in this case, we would like to create a situation where room (A) is one table and room (B) is the table next to it, and both parties can hear each other's conversation to some extent, just like in a party room.
If we adopt the method described by Max, we would have to create 10000 voice channels (100*100) for 100 people.
This is not a solution in terms of bandwidth and resources.
how else are you imagining creating spatialized conversations like this without having to relay that many channels? if you were to mix it town in the server, you won’t be able to spatialize it
you’d basically have to create a personalized mix down in the server and hope that the player doesn’t move abruptly. it would be less bandwidth but a lot of CPU
You can implement this using authoritative servers on Normcore Private, but I don’t really know if that approach is worth doing given this feels less central to the core experience of the title
What we're thinking about is making it function like a party table, as opposed to a regular ROOM.
I don't think it would be practical to calculate spatialisation of all avatar sounds.
We think that only the sounds of people gathered in the same room (party table-like) will be spatialised, and that the adjacent rooms (neighbouring party tables) are some distance away, so only the volume and direction of the sound should be calculated. It is safe to assume that the adjacent neighbour's audio can be taken into account without taking it into account.
Using normcore private authority servers.
Return mixed audio to the normcore in an MCU-like manner, not SFU
It is possible to return the voice to the client as one channel of the SFU
Is this correct?
yes Normcore Private lets you ship your own headless unity builds. You can make one that connects to both rooms, mixes the audio down, and sends it new audio streams to each room
Hi Max.
Thank you.
If you set up a private server for Normcore,
We found out that if we set up a private server for Normcore, we could implement the functionality we wanted.
Can I run the private server on Docker?
Normcore Private comes in a version that Normal hosts, and one that you can host yourself with Kubernetes + Docker
You can also upload game server builds to the version Normcore hosts using Docker
Thank you.
>Normcore private servers are offered in a normal hosted version.
What is the cost of using this?
Also.
>Kubernetes + Docker.
Can I get a Docker img and k8S sample files if I take this configuration?
And where can I find the pricing structure for private server usage?
normcore.io/contact. You'll need to reach out